INDEX

Explanations

words related to positive attributes or qualities

oai_token-act-pair · gpt-3.5-turbo

terms associated with societal criticism and morality

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

olkien

-0.80

ORGE

-0.79

glass

-0.78

akings

-0.73

 Unch

-0.73

ERROR

-0.73

AKING

-0.71

cham

-0.69

BOOK

-0.68

~~~~~~~~~~~~~~~~

-0.67

POSITIVE LOGITS

ity

2.79

ities

2.12

itÃ©

1.45

ITY

1.42

itous

1.37

itarian

1.33

ITIES

1.08

ification

1.08

iated

1.05

ified

1.05

Activations Density 0.049%

words related to positive attributes or qualities

terms associated with societal criticism and morality

No Comments

No Known Activations